Abstract
In this paper, an efficient face recognition algorithm has been presented. First, the face image is extracted to reduce the data dimensions. Next, five important facial components are located using Active Shape Model (ASM). Features from these important components are extracted using Local Directional Pattern (LDP). In the recognition module, Support Vector Machine (SVM) is used to train and tested on the resultant features. The high recognition rate of 97% has been obtained by utilizing a minimum number of features using ORL face database. The results of the proposed technique are substantially accurate.